The overall objective of this proposal is to develop a novel strategy to prime for broad T cell immunity against HIV-1 using eukaryotic expression plasmids (DNA vaccines) delivered to mucosal surfaces by a highly attenuated Shigella flexneri (or alternative attenuated) bacterial delivery vehicle. The induction of HIV-1-specific CD8+ CTL responses, in addition to neutralizing antibody, will be an important feature of an effective HIV-1 vaccine. In order to protect against sexual HIV-1 transmission in humans (where infected mucosal cells may not be readily accessible to parenterally primed memory T cells), it may be necessary to induce T cell immunity both systemically and at mucosal sites. The investigators will use a murine model of respiratory Shigella infection to determine which of several attenuated Shigella strains (impaired in replication or in cell-to-cell spread), carrying DNA vaccine plasmids that encode HIV-1 or SIV subunits, engenders the strongest in vitro CD8+ CTL activity. Responses will be measured against MHC class I-matched recombinant vaccinia-HIV-1- or SIV-subunit infected targets, following intranasal (i.n.) inoculation. The investigators will study whether immune priming with an attenuated Shigella (DNA HIV-1 subunit vaccine) construct at the nasal site engenders CD8+ CTLs that recognize and lyse human cells infected with a divergent strain of HIV-1. This will be done by examining the ability of CD8+ CTL splenocytes recovered from immunized transgenic human HLA-A2.1 mice to recognize human cells that express A2.1 and that are infected with HIV-1 strains represented by the immunogen as well as divergent HIV-1 isolates. Next the investigators will optimize T cell priming for HIV-1 or SIV-specific CD8+ responses at distant mucosal sites relevant to protection against mucosal HIV-1 infection in humans, by immunizing mice i.n., orally, or intravaginally with attenuated bacteria carrying DNA vaccines with or without direct intramuscular injection of the corresponding purified DNA HIV-I subunit vaccine plasmid and recover lymphocytes from genital lymph nodes at various intervals to determine which schedule optimizes anti-HIV-1 CD8+ CTL responses. Finally, if this strategy looks promising, the investigators will determine which route of Shigella (pDVP::SIVgag, env, or pol) inoculation, either p.o. or i.n., primes rhesus monkeys for optimal systemic and/or mucosal CD8+ CTL responses that may protect against an intravaginal challenge with virulent SIV.